Table of contents
Overview
A knowledge graph is a network of real-world entities, such as objects, events, situations, or ideas, and shows how they are related. Most of the time, this information is kept in a graph database and viewed as a graph structure. This is where the term “knowledge graph” comes from.
There are three main components that make up a knowledge graph:
- Nodes
- Edges
- Lables
Any object, place, or person can be a node. An edge defines the relationship between the nodes. Having said that knowledge graphs focus on data that is connected.
Why knowledge graph?
A knowledge graph turns our data into knowledge that a machine can understand. In other words, converting our data into machine-understandable knowledge. Its only purpose is to find hidden insights; it doesn’t serve operational purposes.
Knowledge graph vs. relational database
Purpose or goal
The goal of a knowledge graph is to uncover hidden insights; it doesn’t serve operational purposes. However, a relational database serves both operational and analytics purposes.
Storage
In a knowledge graph, entities and relationships are stored as nodes and edges, respectively, while in a relational database, data is stored in tables as rows and columns. Join queries are used to establish the relationships between the tables.
Schema
Knowledge graphs are schema-free and unstructured, while relational databases have strict schemas. The structure and format of the data are predefined.
Performance
Knowledge graphs are lightning fast, even for big data sets. Relational databases are relatively slower than knowledge graphs.
Maintenance
Knowledge graphs are much easier because they don’t have a schema. Relational databases are hard and often cumbersome because small changes can affect the whole structure.
Comments